Ana Riza F. Quiros

Machine vision for traffic violation detection system through genetic algorithm

This paper presents a machine vision algorithm to detect traffic violations specifically swerving and blocking the pedestrian lane. The proposed solution consists of background difference method, and focuses on the genetic algorithm of the system to detect these violations. The general process is as follows: a capture picture is to be subtracted first by the reference image, then the genetic algorithm is run to find the violator, and finally a display is outputted with the corresponding type of violation. The machine vision traffic violation detection system was found to have an average convergence of about 8 iterations, within an average of less than 300 generations. These results show that the algorithm is well-suited for real time implementation in traffic detection system. Provided the system inputs were captured photos from a CCTV camera, whereas the outputs were cropped pictures of the car that was detected to have such violations mentioned earlier.

A genetic algorithm and artificial neural network-based approach for the machine vision of plate segmentation and character recognition

This paper proposes a genetic-algorithm and neural network-based approach in the optimization of the process of plate segmentation and character recognition respectively in intelligent transportation systems. Upon the detection of the vehicles plate from a captured image, it is necessary that the individual characters in the detected plate are distinguished. After the process of plate recognition, the recognized plate number can be crossed-referenced against a database to correctly identify the vehicles owner and ultimately penalize him for the traffic rule he violated. The segmentation algorithm captures the region of each character in the detected plate using genetic algorithm. After which, each plate character image is mapped against its corresponding sample character image. This is done by feeding sample character images into an artificial neural network and training the network.

Machine vision of traffic state estimation using fuzzy logic

One of the problems encountered by motorists are congested roads. Current technology cannot easily broadcast the information about which roads are heavily congested and which are not to the motorists. As such, planning of the route to take to their destinations is compromised. This paper proposes a fuzzy logic method approach to the estimation of the traffic state of a road. Images from IP cameras installed in different roads can be used to determine the state of the traffic in an area at any point in time. The vehicles within the image are needed to be detected first via edge detection. As the vehicles are detected within the image, so are their position and size with respect to the whole image are obtained. As such, three different parameters namely vehicle density, distance between neighboring vehicles and vehicle sizes can be computed. Using these three parameters, a fuzzy logic system can be created. Three degrees of intensity for each parameter was used, creating 27 rules. The center of gravity method was used to defuzzify the traffic density parameter. Based on the results, the designed algorithm was able to identify six different road images of different traffic states accurately.

Automated traffic violation apprehension system using genetic algorithm and artificial neural network

Developing countries face the problem of crowded and congested roads because of inefficient implementation of traffic rules. Motorists ignore the rules because they are not apprehended and can get away easily. This paper proposes an intelligent traffic system that is able to automatically detect and apprehend traffic violators, specifically motorists who either swerve or block the pedestrian lane. The system is designed by integrating three processes: violation detection, plate localization and plate recognition. The violation detection and plate localization were realized using genetic algorithm while the plate recognition process was performed using an artificial neural network. The recognition of the plate number is highly dependent on the position of the detected vehicle with respect to the camera. Thus, the recognized plate number will only be supplementary information about the violator; the physical attributes of the vehicle which is captured by the violation detection process will be the main information on the violator. Based on the results of 48 images tested, the overall system was able to detect the mentioned violations and to identify the plate number of the vehicles that were detected as traffic violators, with an average accuracy of 90.67%, and program runtime of 1.34 seconds.

Fuzzy logic based vehicular plate character recognition system using image segmentation and scale-invariant feature transform

This paper proposes a vehicle plate optical character recognition method using scale invariant feature transform integrated with image segmentation and fuzzy logic. Image segmentation separates every character in a plate area to get the features of every character obtained. Scale Invariant Feature Transform or SIFT on the other hand, allows the extraction of every feature of each character obtained from the plate. Fuzzy logic analyzes the features obtained from the SIFT algorithm which is proposed to detect the characters correctly. This program used MATLAB to determine the performance of the algorithm. Using the proposed algorithm, it was shown how the algorithm was effective on extracting plate character features as well as recognizing the characters in a given image. Results show that the algorithm has an accuracy of 90.75% and now ready to use for other implementation. This can be incorporated to present optical character recognition system and test its validity and accuracy for practical purposes.

Comparison of Huffman Algorithm and Lempel-Ziv Algorithm for audio, image and text compression

In digital communications, it is necessary to compress the data for a faster and more reliable transmission. As such, the data should undergo source encoding, also known as data compression, which is the process by which data are compressed into a fewer number of bits, before transmission. Also, source encoding is essential to limit file sizes for data storage. Two of the most common and most widely used source encoding techniques are the Huffman Algorithm and Lempel-Ziv Algorithm. The main objective of this research is to identify which technique is better in text, image and audio compression applications. The files for each data type were converted into bit streams using an analog-to-digital converter and pulse code modulation. The bit streams underwent compression through both compression algorithms and the efficiency of each algorithm is quantified by measuring their compression ratio for each data type.

Philippine vehicle plate localization using image thresholding and genetic algorithm

This paper proposes a vehicle plate localization method using genetic algorithm integrated with image thresholding. Image thresholding outputs a value which varies on the time the image is captured. Genetic algorithm on the other hand, executed the license plate region detection of the digital image which depends on the set-level of the image threshold values obtained. Using the proposed algorithm, it was shown how the algorithm was effective on finding the plate location in a given image. Results show that the different parameters tested were successful and converges to a point where the plate locations can be located. The algorithms were tested on an image of a vehicle equipped with a license plate on its frontal view tested on a large number of trials. The genetic algorithm initialized 2000 chromosomes as its initial population and a fixed generations count of 100. It was observed that the time it took for the program to locate the plate is about 3 seconds. Another finding observed is that by varying the initial chromosome count and generation count will lead to longer computation time with increased accuracy. On the contrary, if the initial values were lessened, computation time will be less but the accuracy lessen. Results show that this plate localization technique successfully locates the plate and may be calibrated depending on the time of analysis.

Object locator and collector robotic arm using artificial neural networks

This paper suggests an artificial neural network approach to an object locator and picker. A robotic arm with two joints and a rotating base will function as a pick-and-place machine. The system follows the following constraints: (1) the base of the robotic arm will be situated at a predetermined and fixed position, hence limiting the area at which it can locate and pick an object and; (2) the object will be placed on a flat surface. The span of the robotic arm will determine this area. Also, it should be noted that the arms base and joints can move for limited angles only. The area of interest by the arm will be mapped into grids with coordinates. The inputs to the artificial neural network system will be the coordinates at which the object was positioned. Its outputs will be the angles of each joint of the robotic arm such that it can pick the object at its corresponding position.